Beijing, May 2015
Rasim M. Khaziakhmetov, Victor S. Seleznev, Alexey V. Liseikin, Alexey A. Bryksin, Pavel V. Gromyko
Чистая энергия
Safe operation of the Sayano-Shushenskaya HPP.
The cause of the accident dated 2009/08/17 and lessons learned based on the geophysical Survey of the Siberian Branch of
the Russian Academy of Sciences
Short description of the Sayano-Shushenskaya Hydroelectric Power
Plant (SSHPP) and conclusions of the Governmental Commission
on the origins of the accident dated 2009/08/17
2
Чистая энергия
Sayano-Shushenskaya Hydroelectric Power Plant (SSHPP) is the
largest HPP in Russia with total installed capacity of 6400 MW
situated in Siberian part of Russia
The Governmental Commission was established to investigate the
cause of the accident at the SSHPP dated 17/08/2009 when
all machine hall equipment was destroyed or seriously impacted
and 75 lives were lost.
The main conclusion is that “turbine cover bolts were destroyed
due to ADDITIONAL dynamic forces”
The dam safety was provided.
The safe spillway operation in winter time was provided.
All machine hall equipment was replaced in accordance to
the newly developed safety requirements.
On 2014/12/12 SSHPP returned to the full scale operations
with 10 new units of the same capacity
But what was the source of the “…ADDITIONAL dynamic
forces…”?
A side view of the SSHPP (a photo from Wikimedia, 2010)
Geophysical Survey. Scheme of instrument locations before and
after accident
3
Чистая энергия
The waves propagating
from the operating HPP
and recorded by seismic
stations, are similar to
recordings of the sounds
of the orchestra
Seismic sensors (triangles)
before the accident were
established only in the body of
the dam of Sayano-
Shushenskaya HPP.
After the accident, additional
recording equipment was
installed in the machine hall of
the HPP
Example of relationships between SSHPP’s working regimes and
registered seismic data
4
Чистая энергия
1, 2 – Oscillations amplitudes changes at frequency of 38.1 Hz (vertical channel, dam body (um/s2) and CERR
(1500*um/s2), accordingly). 3 – Change in produced power by Unit#6 (MW).
Oscillations (vibrations) caused by the operation of the hydraulic units are well
recordable by seismic stations even at a distance of several kilometers
Correlation between
operating mode of the
Unit#6 and oscillations
amplitudes changes is
clear and strong even in
cases of idling, ramping
up and down (curve #3)
Water hummer, external impact, strong vibrations?
5
Чистая энергия
It was a lot of versions based on
hypotheses of water hummer or
other shock impact of “…the
ADDITIONAL force...”
Seismological station data
recorded at the time of the
accident indicate that the cause
of the accident is not associated
with shock effects. No indication
of impacts stronger than 10 kJ
was detected.
Seismic sensor is very sensitive -
22 minutes before the accident
the earthquake near Japan was
clearly registered.
Vertical channel data recording (at CERR) at the
moment of accident
Data analyzed from instruments located in the dam of SSHPP
6
Чистая энергия
(a) Recordings of modeled slow movement. (b) Real data from different locations within
SSHPP body
Data on the dam
oscillations at the time of
the accident helped to
determine that the
hydraulic unit#2 rose and
fell slowly enough –
approximately within 11
seconds
Photo of the machine hall destructions after the accident
Spectrogram of seismic recordings prior to the accident (registered
at CERR)
7
Чистая энергия
Changes of oscillations amplitudes with frequencies of 1.55 and 1.85 Hz
(registered at CERR)
In the records of the
Seismic Station
"Cheremushki" there is
no indication that the
amplitude of the
oscillations of hydraulic
units increased prior to
the accident
Oscillations spectra registered on turbine covers (generators under
a load of 640 MW)
8
Чистая энергия
1, 2 – for Unit#9 (new design) and for Unit#3 (old design) accordingly
Oscillations analysis has detected
high frequency vibrations for units
of old design (Unit#3, curve #2).
The turbine cover and the studs
have long been exposed to high
frequency vibrations giving rise to
fatigue defects (despite the fact that
the amplitude of the vibrations was
236 times lower the established
limits)
Conclusions
9
Чистая энергия
Analysis of seismic records obtained by Seismic Station "Cheryomushki" and
stations installed in the dam body revealed that:
1. Cause of the accident was not a shock or hit (water hammer or
“…ADDITIONAL force…”);
2. The was no degradation in vibration conditions of hydraulic units before
the accident;
3. The most probable cause of the accident is the destruction of studs (bolts) of
turbine cover due to the long term influence of high-frequency vibration
(with amplitudes 236 times lower then safety limits) and accumulation of
fatigue defects
4. Before the accident, the main priority was given for monitoring the safety
parameters of the dam. After the accident, we came to the conclusion that
we should not pay less attention to online monitoring of ALL hydraulic
structures, ALL equipment and their mutual influence.
5. The methodology used for remote monitoring of the safety parameters of
the HPP’s equipment can be recommended for use. It is desirable to
formalize it as an industry safe operation standard